@book{MTMT:34832829,
	title = {Fast Electrical Load Classification Using a Dimmer-Based Smart Plug},
	url = {https://m2.mtmt.hu/api/publication/34832829},
	isbn = {978-3-031-37470-8},
	author = {Németh, Dániel and Tornai, Kálmán},
	editor = {Klein, Cornel and Jarke, Matthias and Ploeg, Jeroen and Helfert, Markus and Berns, Karsten and Gusikhin, Oleg},
	publisher = {Springer Nature Switzerland},
	unique-id = {34832829},
	abstract = {As renewable energy resources become a significant source of electricity production, the stable operation of the electrical grid becomes increasingly difficult. Demand-side control of the electrical grid load solves this problem and enables better utilization of renewable energy such as wind or solar power. Adjusting the grid load to meet the renewable production levels requires knowledge about the composition of the grid load as well as the ability to schedule individual loads. We propose a Smart Plug solution capable of accurately classifying the connected electrical load as well as running the Neural Network-based classification on the Smart Plug. The Smart Plug is WiFi-capable allowing wireless measurements as well as remote control of the connected electrical load. We took measurements with the Smart Plug prototype of common household electrical loads and achieved very high accuracy. This accuracy rate can be achieved with on-device measurement and on-device NN inference in less than 2.5 s. Multiple NN-based classification methods and measurements of different amounts of data were examined (measurement profiles).},
	year = {2023},
	pages = {68-89},
	orcid-numbers = {Tornai, Kálmán/0000-0003-1852-0816}
}

@article{MTMT:33864634,
	title = {Wastewater-based modeling, reconstruction, and prediction for COVID-19 outbreaks in Hungary caused by highly immune evasive variants},
	url = {https://m2.mtmt.hu/api/publication/33864634},
	author = {Polcz, Péter and Tornai, Kálmán and Juhász, János and Cserey, György Gábor and Surján, György and Pándics, Tamás and Róka, Eszter and Vargha, Márta and Reguly, István Zoltán and Csikász-Nagy, Attila and Pongor, Sándor and Szederkényi, Gábor},
	doi = {10.1016/j.watres.2023.120098},
	journal-iso = {WATER RES},
	journal = {WATER RESEARCH},
	volume = {241},
	unique-id = {33864634},
	issn = {0043-1354},
	year = {2023},
	eissn = {1879-2448},
	orcid-numbers = {Tornai, Kálmán/0000-0003-1852-0816; Surján, György/0000-0001-5836-7647; Reguly, István Zoltán/0000-0002-4385-4204; Csikász-Nagy, Attila/0000-0002-2919-5601; Szederkényi, Gábor/0000-0003-4199-6089}
}

@article{MTMT:33776658,
	title = {Electrical Load Classification with Open-Set Recognition},
	url = {https://m2.mtmt.hu/api/publication/33776658},
	author = {Németh, Dániel and Tornai, Kálmán},
	doi = {10.3390/en16020800},
	journal-iso = {ENERGIES},
	journal = {ENERGIES},
	volume = {16},
	unique-id = {33776658},
	issn = {1996-1073},
	abstract = {Full utilization of renewable energy resources is a difficult task due to the constantly changing demand-side load of the electrical grid. Demand-side management would solve this crucial problem. To enable demand-side management, knowledge about the composition of the grid load is required, as well as the ability to schedule individual loads. There are proposed Smart Plugs presented in the literature capable of such tasks. The problem, however, is that these methods lack the ability to detect if a previously unseen electrical load is connected. Misclassification of such loads presents a problem for load estimation and scheduling. Open-set recognition methods solve this problem by providing a way to detect samples not belonging to any class used during the training of the classifier. This paper evaluates the novel application of open-set recognition methods to the problem of load classification. Two approaches were examined, and both offer promising results. A Support Vector Machine based approach was first evaluated. The second, more robust method used a modified OpenMax-based algorithm to detect unseen loads.},
	keywords = {Smart grid; Convolutional neural networks; smart home; Smart plug; open-set recognition; electrical load classification},
	year = {2023},
	eissn = {1996-1073},
	orcid-numbers = {Tornai, Kálmán/0000-0003-1852-0816}
}

@article{MTMT:33697412,
	title = {Improving the Performance of Open-Set Recognition with Generated Fake Data},
	url = {https://m2.mtmt.hu/api/publication/33697412},
	author = {Halász, András Pál and Al Hemeary, Nawar and Daubner, Lóránt Szabolcs and Zsedrovits, Tamás and Tornai, Kálmán},
	doi = {10.3390/electronics12061311},
	journal = {ELECTRONICS (SWITZ)},
	volume = {12},
	unique-id = {33697412},
	abstract = {Open-set recognition models, in addition to generalizing to unseen instances of known categories, have to identify samples of unknown classes during the training phase. The main reason the latter is much more complicated is that there is very little or no information about the properties of these unknown classes. There are methodologies available to handle the unknowns. One possible method is to construct models for them by using generated inputs labeled as unknown. Generative adversarial networks are frequently deployed to generate synthetic samples representing unknown classes to create better models for known classes. In this paper, we introduce a novel approach to improve the accuracy of recognition methods while reducing the time complexity. Instead of generating synthetic input data to train neural networks, feature vectors are generated using the output of a hidden layer. This approach results in a less complex structure for the neural network representation of the classes. A distance-based classifier implemented by a convolutional neural network is used in our implementation. Our solution’s open-set detection performance reaches an AUC value of 0.839 on the CIFAR-10 dataset, while the closed-set accuracy is 91.4%, the highest among the open-set recognition methods. The generator and discriminator networks are much smaller when generating synthetic inner features. There is no need to run these samples through the first part of the classifier with the convolutional layers. Hence, this solution not only gives better performance than generating samples in the input space but also makes it less expensive in terms of computational complexity.},
	keywords = {GAN; Distance-based classification; open-set recognition; sample generation},
	year = {2023},
	eissn = {2079-9292},
	pages = {1311-6},
	orcid-numbers = {Zsedrovits, Tamás/0000-0003-0768-1171; Tornai, Kálmán/0000-0003-1852-0816}
}

@book{MTMT:33704221,
	title = {The Design and Utilisation of PanSim, a Portable Pandemic Simulator},
	url = {https://m2.mtmt.hu/api/publication/33704221},
	author = {Keömley-Horváth, Bence and Horváth, Gergely and Polcz, Péter and Siklósi, Bálint and Tornai, Kálmán and Juhász, János and Szederkényi, Gábor and Cserey, György Gábor and Csikász-Nagy, Attila and Reguly, István Zoltán},
	doi = {10.1109/CIW-IUS56691.2022.00006},
	publisher = {Institute of Electrical and Electronics Engineers},
	unique-id = {33704221},
	year = {2022},
	pages = {1-9},
	orcid-numbers = {Tornai, Kálmán/0000-0003-1852-0816; Szederkényi, Gábor/0000-0003-4199-6089; Csikász-Nagy, Attila/0000-0002-2919-5601; Reguly, István Zoltán/0000-0002-4385-4204}
}

@inproceedings{MTMT:33103187,
	title = {Detecting Unknown Electrical Loads Using Open Set Recognition},
	url = {https://m2.mtmt.hu/api/publication/33103187},
	author = {Németh, Dániel and Tornai, Kálmán},
	booktitle = {2022 IEEE 10th International Conference on Smart Energy Grid Engineering (SEGE)},
	doi = {10.1109/SEGE55279.2022.9889770},
	unique-id = {33103187},
	year = {2022},
	pages = {7-11},
	orcid-numbers = {Tornai, Kálmán/0000-0003-1852-0816}
}

@inproceedings{MTMT:32803859,
	title = {SP4LC: A Method for Recognizing Power Consumers in a Smart Plug},
	url = {https://m2.mtmt.hu/api/publication/32803859},
	author = {Németh, Dániel and Tornai, Kálmán},
	booktitle = {Proceedings of the 11th International Conference on Smart Cities and Green ICT Systems},
	doi = {10.5220/0010982800003203},
	unique-id = {32803859},
	year = {2022},
	pages = {69-77},
	orcid-numbers = {Tornai, Kálmán/0000-0003-1852-0816}
}

@article{MTMT:32784092,
	title = {Integral representation method based efficient rule optimizing framework for anti-money laundering},
	url = {https://m2.mtmt.hu/api/publication/32784092},
	author = {Badics, Tamás and Hajtó, Dániel and Tornai, Kálmán and Kiss, Levente and Reguly, István Zoltán and Pesti, István and Sváb, Péter and Cserey, György Gábor},
	doi = {10.1108/JMLC-12-2021-0137},
	journal-iso = {J MONEY LAUNDER CONT},
	journal = {JOURNAL OF MONEY LAUNDERING CONTROL},
	volume = {26},
	unique-id = {32784092},
	issn = {1368-5201},
	abstract = {Purpose This paper aims to introduce a framework for optimizing rule-based anti-money laundering systems with a clear economic interpretation, and the authors introduce the integral representation method. Design/methodology/approach By using a microeconomic model, the authors reformulate the threshold optimization problem as a decision problem to gain insights from economics regarding the main properties of the optimum. The authors used algorithmic considerations to find an efficient implementation by using a kind of weak mode estimate of the distribution and the authors extend this approach to classes of alerts or cases. Findings The method provides a new and efficient alternative for the sampling method or the multidimensional optimization technique described in the literature to decrease the bias emanating from multiple alerts by smoothing the number of alerts across classes in the optimum and decrease the overlapping between scenarios at the case level. Using the method for real bank data, the authors were able to decrease the number of false positives cases by about 18% while retaining almost 98% of the true-positive cases. Research limitations/implications The model assumes that alerts from different scenarios are indifferent to the bank. To include scenario-specific preferences or constraints demands further research. Originality/value The new framework presented in the paper is a flexible extension of the usual above-the-line method, which makes it possible to include bank preferences and use the parallelization capabilities of modern processors.},
	year = {2022},
	eissn = {1758-7808},
	pages = {290-308},
	orcid-numbers = {Tornai, Kálmán/0000-0003-1852-0816; Reguly, István Zoltán/0000-0002-4385-4204}
}

@article{MTMT:32574366,
	title = {Microsimulation based quantitative analysis of COVID-19 management strategies},
	url = {https://m2.mtmt.hu/api/publication/32574366},
	author = {Reguly, István Zoltán and Csercsik, Dávid and Juhász, János and Tornai, Kálmán and Bujtár, Zsófia and Horváth, Gergely and Keömley-Horváth, Bence and Kós, Tamás and Cserey, György Gábor and Iván, Kristóf and Pongor, Sándor and Szederkényi, Gábor and Röst, Gergely and Csikász-Nagy, Attila},
	doi = {10.1371/journal.pcbi.1009693},
	journal-iso = {PLOS COMPUT BIOL},
	journal = {PLOS COMPUTATIONAL BIOLOGY},
	volume = {18},
	unique-id = {32574366},
	issn = {1553-734X},
	year = {2022},
	eissn = {1553-7358},
	orcid-numbers = {Reguly, István Zoltán/0000-0002-4385-4204; Tornai, Kálmán/0000-0003-1852-0816; Kós, Tamás/0000-0001-5105-4699; Iván, Kristóf/0000-0003-3637-3979; Szederkényi, Gábor/0000-0003-4199-6089; Röst, Gergely/0000-0001-9476-3284; Csikász-Nagy, Attila/0000-0002-2919-5601}
}

@article{MTMT:31674480,
	title = {Intelligent Sensor Data Analysis form Smart Systems},
	url = {https://m2.mtmt.hu/api/publication/31674480},
	author = {Márkos, Zsolt and Tornai, Kálmán},
	journal-iso = {JEDLIK LABOR REP},
	journal = {JEDLIK LABORATORIES REPORTS},
	volume = {2020},
	unique-id = {31674480},
	issn = {2064-3942},
	year = {2020},
	pages = {40-44},
	orcid-numbers = {Tornai, Kálmán/0000-0003-1852-0816}
}